The Average Coding Length of Huffman Coding Based Signal Processing and Its Application in Fault Severity Recognition

Abstract

The transient impact components in vibration signal, which are the major information for bearing fault severity recognition, are often interfered with by ambient noise. Meanwhile, for bearing fault severity recognition, the frequency band selection methods which are employed to pre-process the contaminated vibration signal only select the partial frequency band of the vibration signal and cause information loss of other frequency band. Aiming at this issue, this paper proposes a novel fault severity recognition method based on Huffman coding, which can retain all the information of the frequency band, and is applied for the first time to bearing fault severity recognition. Specifically, the average coding length of Huffman coding (ACLHC) of the original vibration signal is first calculated to reduce the noise and highlight the impact components of the signal. Then, the ACLHC is encoded by symbolic aggregate approximation (SAX) to reflect the modulation information of bearing. Finally, the Lempel‑Ziv indicator (LZ indicator) of the symbol sequence is calculated to reflect the fault severity. The proposed method is verified by the bearing datasets under different working conditions. Compared with the methods based on frequency band selection, the proposed method effectively recognizes the fault severity of bearing for more working conditions.

Fault diagnosis of various rotating equipment using machine learning approaches – A review

“…93 A frequency matching linear transform technique is reported for bearing fault detection under variable rotating speeds 94 and Huffman coding technique is also used to identify bearing defect severity. 95…”

Fault Diagnosis of Rotating Machine

“…In the first paper, Duan et al [39] improved the fault detection rate of a general rolling bearing by combining the local mean decomposition (LMD) and the ratio correction methods. Then, Cui et al [40] discussed the diagnosis of multiple defects in a rolling bearing via vibration analysis; Shi et al [41] reported a frequency matching linear transform technique for bearing fault detection under variable rotating speeds; moreover, Yin et al [42] proposed a Huffman coding technique to identify bearing defect severity. By what follows, artificially intelligent techniques, including ensemble learning [43] and deep learning [44,45] were developed to detect bearing faults.…”

Convolutional neural network with Huffman pooling for handling data with insufficient categories: A novel method for anomaly detection and fault diagnosis

“…It is found that Huffman coding can effectively reduce the noise of the vibration signal of the rotor equipment and enhance the impact characteristics of the vibration signal in our previous study. 35 Considering that some effective feature information will be lost in the preprocessing of vibration signal, we try to combine Huffman coding with CNN, and conduct Huffman coding on the data stream during the CNN feature extraction process. In view of CNN's weak ability to identify unknown test samples, we further enhanced CNN's ability to describe normal data by increasing the distribution difference between classes of known training data to isolate the characteristic space of normal data.…”